Super-resolution reconstruction method based on dirac residual deep neural network

Abstract

The invention discloses a super-resolution reconstruction method based on a dirac residual deep neural network. The super-resolution reconstruction method comprises the steps: a network inputs a low-resolution image, learns image features through a plurality of dirac parameterized residual blocks, and employs sub-pixel convolution to reconstruct a high-resolution image; the network is divided intoan upper part and a lower part, and the upper part obtains high-frequency features of the LR through the deep dirac residual network, and then carries out reconstruction through sub-pixel convolution; the lower part reconstructs an image directly by sub-pixel convolution of low frequency features of the LR; and a reconstructed HR image is output by combining the two reconstruction structures. According to the super-resolution reconstruction method, the residual layer is improved through weight parameterization; the convolution feature dimension before the activation function Relu is reduced;and the convolution feature dimension after the activation function is increased. Meanwhile, the convolutional features of the input image and the features of residual network learning are combined for reconstruction, and a better SR effect is obtained under the condition that the network depths are the same.

Description

technical field [0001] The invention relates to a super-resolution reconstruction method based on a dirac residual deep neural network, and belongs to the technical field of image reconstruction. Background technique [0002] Super-resolution reconstruction (SR) refers to the estimation of high-resolution (high-resolution, HR) images or videos obtained by observing one or more low-resolution (low-resolution, LR) images of the same scene. . SR breaks through the limitations of imaging system hardware through digital image processing and machine learning methods to obtain images with higher spatial resolution and more detailed information. It is currently an efficient and low-cost technical means to obtain high-definition images. [0003] SR can be divided into three categories according to different input and output methods, namely single-input single-output (SISO), multiple-input single-output (MISO) and multiple-input multiple-output (MIMO), among which MIMO belongs to vid...

AI Technical Summary

Problems solved by technology

Each residual block is an activation function added to the two convolutional layers, and the output of the second convolution is added to the input of the first convolution as the output of the residual block. When the network depth increases, The reconstruction effect is not improved well

Images